From: Sergey M Date: Sat, 31 Oct 2015 18:15:21 +0000 (+0000) Subject: Merge pull request #7296 from jaimeMF/xml_attrib_unicode X-Git-Url: http://git.bitcoin.ninja/index.cgi?a=commitdiff_plain;h=30eecc6a044d4070d0e8ea4a6e0637867f0d3a28;hp=ae37338e681319a28d98dc551253d9fa1830969a;p=youtube-dl Merge pull request #7296 from jaimeMF/xml_attrib_unicode Use a wrapper around xml.etree.ElementTree.fromstring in python 2.x (… --- diff --git a/test/test_utils.py b/test/test_utils.py index a9e0fed7e..3298315d2 100644 --- a/test/test_utils.py +++ b/test/test_utils.py @@ -278,9 +278,16 @@ class TestUtil(unittest.TestCase): p = xml.etree.ElementTree.SubElement(div, 'p') p.text = 'Foo' self.assertEqual(xpath_element(doc, 'div/p'), p) + self.assertEqual(xpath_element(doc, ['div/p']), p) + self.assertEqual(xpath_element(doc, ['div/bar', 'div/p']), p) self.assertEqual(xpath_element(doc, 'div/bar', default='default'), 'default') + self.assertEqual(xpath_element(doc, ['div/bar'], default='default'), 'default') self.assertTrue(xpath_element(doc, 'div/bar') is None) + self.assertTrue(xpath_element(doc, ['div/bar']) is None) + self.assertTrue(xpath_element(doc, ['div/bar'], 'div/baz') is None) self.assertRaises(ExtractorError, xpath_element, doc, 'div/bar', fatal=True) + self.assertRaises(ExtractorError, xpath_element, doc, ['div/bar'], fatal=True) + self.assertRaises(ExtractorError, xpath_element, doc, ['div/bar', 'div/baz'], fatal=True) def test_xpath_text(self): testxml = ''' @@ -428,6 +435,8 @@ class TestUtil(unittest.TestCase): self.assertEqual(parse_iso8601('2014-03-23T22:04:26+0000'), 1395612266) self.assertEqual(parse_iso8601('2014-03-23T22:04:26Z'), 1395612266) self.assertEqual(parse_iso8601('2014-03-23T22:04:26.1234Z'), 1395612266) + self.assertEqual(parse_iso8601('2015-09-29T08:27:31.727'), 1443515251) + self.assertEqual(parse_iso8601('2015-09-29T08-27-31.727'), None) def test_strip_jsonp(self): stripped = strip_jsonp('cb ([ {"id":"532cb",\n\n\n"x":\n3}\n]\n);') diff --git a/youtube_dl/extractor/__init__.py b/youtube_dl/extractor/__init__.py index 6318ac4a2..f98e6487e 100644 --- a/youtube_dl/extractor/__init__.py +++ b/youtube_dl/extractor/__init__.py @@ -90,6 +90,7 @@ from .cliphunter import CliphunterIE from .clipsyndicate import ClipsyndicateIE from .cloudy import CloudyIE from .clubic import ClubicIE +from .clyp import ClypIE from .cmt import CMTIE from .cnet import CNETIE from .cnn import ( diff --git a/youtube_dl/extractor/anitube.py b/youtube_dl/extractor/anitube.py index 31f0d417c..23f942ae2 100644 --- a/youtube_dl/extractor/anitube.py +++ b/youtube_dl/extractor/anitube.py @@ -26,8 +26,8 @@ class AnitubeIE(InfoExtractor): video_id = mobj.group('id') webpage = self._download_webpage(url, video_id) - key = self._html_search_regex( - r'http://www\.anitube\.se/embed/([A-Za-z0-9_-]*)', webpage, 'key') + key = self._search_regex( + r'src=["\']https?://[^/]+/embed/([A-Za-z0-9_-]+)', webpage, 'key') config_xml = self._download_xml( 'http://www.anitube.se/nuevo/econfig.php?key=%s' % key, key) diff --git a/youtube_dl/extractor/clyp.py b/youtube_dl/extractor/clyp.py new file mode 100644 index 000000000..57e643799 --- /dev/null +++ b/youtube_dl/extractor/clyp.py @@ -0,0 +1,57 @@ +from __future__ import unicode_literals + +from .common import InfoExtractor +from ..utils import ( + float_or_none, + parse_iso8601, +) + + +class ClypIE(InfoExtractor): + _VALID_URL = r'https?://(?:www\.)?clyp\.it/(?P[a-z0-9]+)' + _TEST = { + 'url': 'https://clyp.it/ojz2wfah', + 'md5': '1d4961036c41247ecfdcc439c0cddcbb', + 'info_dict': { + 'id': 'ojz2wfah', + 'ext': 'mp3', + 'title': 'Krisson80 - bits wip wip', + 'description': '#Krisson80BitsWipWip #chiptune\n#wip', + 'duration': 263.21, + 'timestamp': 1443515251, + 'upload_date': '20150929', + }, + } + + def _real_extract(self, url): + audio_id = self._match_id(url) + + metadata = self._download_json( + 'https://api.clyp.it/%s' % audio_id, audio_id) + + formats = [] + for secure in ('', 'Secure'): + for ext in ('Ogg', 'Mp3'): + format_id = '%s%s' % (secure, ext) + format_url = metadata.get('%sUrl' % format_id) + if format_url: + formats.append({ + 'url': format_url, + 'format_id': format_id, + 'vcodec': 'none', + }) + self._sort_formats(formats) + + title = metadata['Title'] + description = metadata.get('Description') + duration = float_or_none(metadata.get('Duration')) + timestamp = parse_iso8601(metadata.get('DateCreated')) + + return { + 'id': audio_id, + 'title': title, + 'description': description, + 'duration': duration, + 'timestamp': timestamp, + 'formats': formats, + } diff --git a/youtube_dl/extractor/common.py b/youtube_dl/extractor/common.py index 52523d7b2..5e263f8b5 100644 --- a/youtube_dl/extractor/common.py +++ b/youtube_dl/extractor/common.py @@ -310,11 +310,11 @@ class InfoExtractor(object): @classmethod def ie_key(cls): """A string for getting the InfoExtractor with get_info_extractor""" - return cls.__name__[:-2] + return compat_str(cls.__name__[:-2]) @property def IE_NAME(self): - return type(self).__name__[:-2] + return compat_str(type(self).__name__[:-2]) def _request_webpage(self, url_or_request, video_id, note=None, errnote=None, fatal=True): """ Returns the response handle """ @@ -943,13 +943,14 @@ class InfoExtractor(object): if re.match(r'^https?://', u) else compat_urlparse.urljoin(m3u8_url, u)) - m3u8_doc, urlh = self._download_webpage_handle( + res = self._download_webpage_handle( m3u8_url, video_id, note=note or 'Downloading m3u8 information', errnote=errnote or 'Failed to download m3u8 information', fatal=fatal) - if m3u8_doc is False: - return m3u8_doc + if res is False: + return res + m3u8_doc, urlh = res m3u8_url = urlh.geturl() last_info = None last_media = None diff --git a/youtube_dl/extractor/francetv.py b/youtube_dl/extractor/francetv.py index 129984a5f..8e60cf60f 100644 --- a/youtube_dl/extractor/francetv.py +++ b/youtube_dl/extractor/francetv.py @@ -83,6 +83,14 @@ class FranceTVBaseInfoExtractor(InfoExtractor): if subtitle: title += ' - %s' % subtitle + subtitles = {} + subtitles_list = [{ + 'url': subformat['url'], + 'ext': subformat.get('format'), + } for subformat in info.get('subtitles', []) if subformat.get('url')] + if subtitles_list: + subtitles['fr'] = subtitles_list + return { 'id': video_id, 'title': title, @@ -91,20 +99,27 @@ class FranceTVBaseInfoExtractor(InfoExtractor): 'duration': int_or_none(info.get('real_duration')) or parse_duration(info['duree']), 'timestamp': int_or_none(info['diffusion']['timestamp']), 'formats': formats, + 'subtitles': subtitles, } class PluzzIE(FranceTVBaseInfoExtractor): IE_NAME = 'pluzz.francetv.fr' - _VALID_URL = r'https?://pluzz\.francetv\.fr/videos/(.*?)\.html' + _VALID_URL = r'https?://(?:m\.)?pluzz\.francetv\.fr/videos/(?P.+?)\.html' # Can't use tests, videos expire in 7 days def _real_extract(self, url): - title = re.match(self._VALID_URL, url).group(1) - webpage = self._download_webpage(url, title) - video_id = self._search_regex( - r'data-diffusion="(\d+)"', webpage, 'ID') + display_id = self._match_id(url) + + webpage = self._download_webpage(url, display_id) + + video_id = self._html_search_meta( + 'id_video', webpage, 'video id', default=None) + if not video_id: + video_id = self._search_regex( + r'data-diffusion=["\'](\d+)', webpage, 'video id') + return self._extract_video(video_id, 'Pluzz') @@ -120,6 +135,9 @@ class FranceTvInfoIE(FranceTVBaseInfoExtractor): 'title': 'Soir 3', 'upload_date': '20130826', 'timestamp': 1377548400, + 'subtitles': { + 'fr': 'mincount:2', + }, }, }, { 'url': 'http://www.francetvinfo.fr/elections/europeennes/direct-europeennes-regardez-le-debat-entre-les-candidats-a-la-presidence-de-la-commission_600639.html', diff --git a/youtube_dl/extractor/generic.py b/youtube_dl/extractor/generic.py index 1de96b268..ee5419f51 100644 --- a/youtube_dl/extractor/generic.py +++ b/youtube_dl/extractor/generic.py @@ -141,6 +141,7 @@ class GenericIE(InfoExtractor): 'ext': 'mp4', 'title': 'Automatics, robotics and biocybernetics', 'description': 'md5:815fc1deb6b3a2bff99de2d5325be482', + 'upload_date': '20130627', 'formats': 'mincount:16', 'subtitles': 'mincount:1', }, diff --git a/youtube_dl/extractor/mdr.py b/youtube_dl/extractor/mdr.py index fc7499958..88334889e 100644 --- a/youtube_dl/extractor/mdr.py +++ b/youtube_dl/extractor/mdr.py @@ -1,64 +1,169 @@ +# coding: utf-8 from __future__ import unicode_literals -import re - from .common import InfoExtractor +from ..compat import compat_urlparse +from ..utils import ( + determine_ext, + int_or_none, + parse_duration, + parse_iso8601, + xpath_text, +) class MDRIE(InfoExtractor): - _VALID_URL = r'^(?Phttps?://(?:www\.)?mdr\.de)/(?:.*)/(?Pvideo|audio)(?P[^/_]+)(?:_|\.html)' + IE_DESC = 'MDR.DE and KiKA' + _VALID_URL = r'https?://(?:www\.)?(?:mdr|kika)\.de/(?:.*)/[a-z]+(?P\d+)(?:_.+?)?\.html' - # No tests, MDR regularily deletes its videos - _TEST = { + _TESTS = [{ + # MDR regularily deletes its videos 'url': 'http://www.mdr.de/fakt/video189002.html', 'only_matching': True, - } + }, { + # audio + 'url': 'http://www.mdr.de/kultur/audio1312272_zc-15948bad_zs-86171fdd.html', + 'md5': '64c4ee50f0a791deb9479cd7bbe9d2fa', + 'info_dict': { + 'id': '1312272', + 'ext': 'mp3', + 'title': 'Feuilleton vom 30. Oktober 2015', + 'duration': 250, + 'uploader': 'MITTELDEUTSCHER RUNDFUNK', + }, + }, { + 'url': 'http://www.kika.de/baumhaus/videos/video19636.html', + 'md5': '4930515e36b06c111213e80d1e4aad0e', + 'info_dict': { + 'id': '19636', + 'ext': 'mp4', + 'title': 'Baumhaus vom 30. Oktober 2015', + 'duration': 134, + 'uploader': 'KIKA', + }, + }, { + 'url': 'http://www.kika.de/sendungen/einzelsendungen/weihnachtsprogramm/videos/video8182.html', + 'md5': '5fe9c4dd7d71e3b238f04b8fdd588357', + 'info_dict': { + 'id': '8182', + 'ext': 'mp4', + 'title': 'Beutolomäus und der geheime Weihnachtswunsch', + 'description': 'md5:b69d32d7b2c55cbe86945ab309d39bbd', + 'timestamp': 1419047100, + 'upload_date': '20141220', + 'duration': 4628, + 'uploader': 'KIKA', + }, + }, { + 'url': 'http://www.kika.de/baumhaus/sendungen/video19636_zc-fea7f8a0_zs-4bf89c60.html', + 'only_matching': True, + }, { + 'url': 'http://www.kika.de/sendungen/einzelsendungen/weihnachtsprogramm/einzelsendung2534.html', + 'only_matching': True, + }] def _real_extract(self, url): - m = re.match(self._VALID_URL, url) - video_id = m.group('video_id') - domain = m.group('domain') + video_id = self._match_id(url) + + webpage = self._download_webpage(url, video_id) + + data_url = self._search_regex( + r'dataURL\s*:\s*(["\'])(?P/.+/(?:video|audio)[0-9]+-avCustom\.xml)\1', + webpage, 'data url', group='url') - # determine title and media streams from webpage - html = self._download_webpage(url, video_id) + doc = self._download_xml( + compat_urlparse.urljoin(url, data_url), video_id) - title = self._html_search_regex(r'(.*?)', html, 'title') - xmlurl = self._search_regex( - r'dataURL:\'(/(?:.+)/(?:video|audio)[0-9]+-avCustom.xml)', html, 'XML URL') + title = xpath_text(doc, ['./title', './broadcast/broadcastName'], 'title', fatal=True) - doc = self._download_xml(domain + xmlurl, video_id) formats = [] - for a in doc.findall('./assets/asset'): - url_el = a.find('./progressiveDownloadUrl') - if url_el is None: - continue - abr = int(a.find('bitrateAudio').text) // 1000 - media_type = a.find('mediaType').text - format = { - 'abr': abr, - 'filesize': int(a.find('fileSize').text), - 'url': url_el.text, - } - - vbr_el = a.find('bitrateVideo') - if vbr_el is None: - format.update({ - 'vcodec': 'none', - 'format_id': '%s-%d' % (media_type, abr), - }) - else: - vbr = int(vbr_el.text) // 1000 - format.update({ - 'vbr': vbr, - 'width': int(a.find('frameWidth').text), - 'height': int(a.find('frameHeight').text), - 'format_id': '%s-%d' % (media_type, vbr), - }) - formats.append(format) + processed_urls = [] + for asset in doc.findall('./assets/asset'): + for source in ( + 'progressiveDownload', + 'dynamicHttpStreamingRedirector', + 'adaptiveHttpStreamingRedirector'): + url_el = asset.find('./%sUrl' % source) + if url_el is None: + continue + + video_url = url_el.text + if video_url in processed_urls: + continue + + processed_urls.append(video_url) + + vbr = int_or_none(xpath_text(asset, './bitrateVideo', 'vbr'), 1000) + abr = int_or_none(xpath_text(asset, './bitrateAudio', 'abr'), 1000) + + ext = determine_ext(url_el.text) + if ext == 'm3u8': + url_formats = self._extract_m3u8_formats( + video_url, video_id, 'mp4', entry_protocol='m3u8_native', + preference=0, m3u8_id='HLS', fatal=False) + elif ext == 'f4m': + url_formats = self._extract_f4m_formats( + video_url + '?hdcore=3.7.0&plugin=aasp-3.7.0.39.44', video_id, + preference=0, f4m_id='HDS', fatal=False) + else: + media_type = xpath_text(asset, './mediaType', 'media type', default='MP4') + vbr = int_or_none(xpath_text(asset, './bitrateVideo', 'vbr'), 1000) + abr = int_or_none(xpath_text(asset, './bitrateAudio', 'abr'), 1000) + filesize = int_or_none(xpath_text(asset, './fileSize', 'file size')) + + f = { + 'url': video_url, + 'format_id': '%s-%d' % (media_type, vbr or abr), + 'filesize': filesize, + 'abr': abr, + 'preference': 1, + } + + if vbr: + width = int_or_none(xpath_text(asset, './frameWidth', 'width')) + height = int_or_none(xpath_text(asset, './frameHeight', 'height')) + f.update({ + 'vbr': vbr, + 'width': width, + 'height': height, + }) + + url_formats = [f] + + if not url_formats: + continue + + if not vbr: + for f in url_formats: + abr = f.get('tbr') or abr + if 'tbr' in f: + del f['tbr'] + f.update({ + 'abr': abr, + 'vcodec': 'none', + }) + + formats.extend(url_formats) + self._sort_formats(formats) + description = xpath_text(doc, './broadcast/broadcastDescription', 'description') + timestamp = parse_iso8601( + xpath_text( + doc, [ + './broadcast/broadcastDate', + './broadcast/broadcastStartDate', + './broadcast/broadcastEndDate'], + 'timestamp', default=None)) + duration = parse_duration(xpath_text(doc, './duration', 'duration')) + uploader = xpath_text(doc, './rights', 'uploader') + return { 'id': video_id, 'title': title, + 'description': description, + 'timestamp': timestamp, + 'duration': duration, + 'uploader': uploader, 'formats': formats, } diff --git a/youtube_dl/extractor/moniker.py b/youtube_dl/extractor/moniker.py index 69e4bcd1a..7c0c4e50e 100644 --- a/youtube_dl/extractor/moniker.py +++ b/youtube_dl/extractor/moniker.py @@ -17,7 +17,7 @@ from ..utils import ( class MonikerIE(InfoExtractor): IE_DESC = 'allmyvideos.net and vidspot.net' - _VALID_URL = r'https?://(?:www\.)?(?:allmyvideos|vidspot)\.net/(?P[a-zA-Z0-9_-]+)' + _VALID_URL = r'https?://(?:www\.)?(?:allmyvideos|vidspot)\.net/(?:(?:2|v)/v-)?(?P[a-zA-Z0-9_-]+)' _TESTS = [{ 'url': 'http://allmyvideos.net/jih3nce3x6wn', @@ -46,6 +46,18 @@ class MonikerIE(InfoExtractor): }, { 'url': 'https://www.vidspot.net/l2ngsmhs8ci5', 'only_matching': True, + }, { + 'url': 'http://vidspot.net/2/v-ywDf99', + 'md5': '5f8254ce12df30479428b0152fb8e7ba', + 'info_dict': { + 'id': 'ywDf99', + 'ext': 'mp4', + 'title': 'IL FAIT LE MALIN EN PORSHE CAYENNE ( mais pas pour longtemps)', + 'description': 'IL FAIT LE MALIN EN PORSHE CAYENNE.', + }, + }, { + 'url': 'http://allmyvideos.net/v/v-HXZm5t', + 'only_matching': True, }] def _real_extract(self, url): @@ -64,18 +76,30 @@ class MonikerIE(InfoExtractor): raise ExtractorError( '%s returned error: %s' % (self.IE_NAME, error), expected=True) - fields = re.findall(r'type="hidden" name="(.+?)"\s* value="?(.+?)">', orig_webpage) - data = dict(fields) + builtin_url = self._search_regex( + r']+src=(["\'])(?P.+?/builtin-.+?)\1', + orig_webpage, 'builtin URL', default=None, group='url') - post = compat_urllib_parse.urlencode(data) - headers = { - b'Content-Type': b'application/x-www-form-urlencoded', - } - req = compat_urllib_request.Request(url, post, headers) - webpage = self._download_webpage( - req, video_id, note='Downloading video page ...') + if builtin_url: + req = compat_urllib_request.Request(builtin_url) + req.add_header('Referer', url) + webpage = self._download_webpage(req, video_id, 'Downloading builtin page') + title = self._og_search_title(orig_webpage).strip() + description = self._og_search_description(orig_webpage).strip() + else: + fields = re.findall(r'type="hidden" name="(.+?)"\s* value="?(.+?)">', orig_webpage) + data = dict(fields) + + post = compat_urllib_parse.urlencode(data) + headers = { + b'Content-Type': b'application/x-www-form-urlencoded', + } + req = compat_urllib_request.Request(url, post, headers) + webpage = self._download_webpage( + req, video_id, note='Downloading video page ...') - title = os.path.splitext(data['fname'])[0] + title = os.path.splitext(data['fname'])[0] + description = None # Could be several links with different quality links = re.findall(r'"file" : "?(.+?)",', webpage) @@ -89,5 +113,6 @@ class MonikerIE(InfoExtractor): return { 'id': video_id, 'title': title, + 'description': description, 'formats': formats, } diff --git a/youtube_dl/extractor/vidme.py b/youtube_dl/extractor/vidme.py index eb5cde761..3d63ed4f0 100644 --- a/youtube_dl/extractor/vidme.py +++ b/youtube_dl/extractor/vidme.py @@ -101,6 +101,10 @@ class VidmeIE(InfoExtractor): # suspended 'url': 'https://vid.me/Ox3G', 'only_matching': True, + }, { + # deleted + 'url': 'https://vid.me/KTPm', + 'only_matching': True, }, { # no formats in the API response 'url': 'https://vid.me/e5g', @@ -143,6 +147,11 @@ class VidmeIE(InfoExtractor): video = response['video'] + if video.get('state') == 'deleted': + raise ExtractorError( + 'Vidme said: Sorry, this video has been deleted.', + expected=True) + if video.get('state') in ('user-disabled', 'suspended'): raise ExtractorError( 'Vidme said: This video has been suspended either due to a copyright claim, ' diff --git a/youtube_dl/utils.py b/youtube_dl/utils.py index c761ea22a..efd5f4ae1 100644 --- a/youtube_dl/utils.py +++ b/youtube_dl/utils.py @@ -179,10 +179,19 @@ def xpath_with_ns(path, ns_map): def xpath_element(node, xpath, name=None, fatal=False, default=NO_DEFAULT): - if sys.version_info < (2, 7): # Crazy 2.6 - xpath = xpath.encode('ascii') + def _find_xpath(xpath): + if sys.version_info < (2, 7): # Crazy 2.6 + xpath = xpath.encode('ascii') + return node.find(xpath) + + if isinstance(xpath, (str, compat_str)): + n = _find_xpath(xpath) + else: + for xp in xpath: + n = _find_xpath(xp) + if n is not None: + break - n = node.find(xpath) if n is None: if default is not NO_DEFAULT: return default @@ -815,9 +824,11 @@ def parse_iso8601(date_str, delimiter='T', timezone=None): if date_str is None: return None + date_str = re.sub(r'\.[0-9]+', '', date_str) + if timezone is None: m = re.search( - r'(\.[0-9]+)?(?:Z$| ?(?P\+|-)(?P[0-9]{2}):?(?P[0-9]{2})$)', + r'(?:Z$| ?(?P\+|-)(?P[0-9]{2}):?(?P[0-9]{2})$)', date_str) if not m: timezone = datetime.timedelta() @@ -830,9 +841,12 @@ def parse_iso8601(date_str, delimiter='T', timezone=None): timezone = datetime.timedelta( hours=sign * int(m.group('hours')), minutes=sign * int(m.group('minutes'))) - date_format = '%Y-%m-%d{0}%H:%M:%S'.format(delimiter) - dt = datetime.datetime.strptime(date_str, date_format) - timezone - return calendar.timegm(dt.timetuple()) + try: + date_format = '%Y-%m-%d{0}%H:%M:%S'.format(delimiter) + dt = datetime.datetime.strptime(date_str, date_format) - timezone + return calendar.timegm(dt.timetuple()) + except ValueError: + pass def unified_strdate(date_str, day_first=True): @@ -897,7 +911,7 @@ def unified_strdate(date_str, day_first=True): timetuple = email.utils.parsedate_tz(date_str) if timetuple: upload_date = datetime.datetime(*timetuple[:6]).strftime('%Y%m%d') - return upload_date + return compat_str(upload_date) def determine_ext(url, default_ext='unknown_video'):